review-pipeline

Reconcile code, test, and security findings into a go/no-go review decision.

2|Updated May 3, 2026
One-click install
npx skills add https://github.com/Rosetears520/aili-workflows --skill review-pipeline-rosetears520
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-pipeline
Source: https://github.com/Rosetears520/aili-workflows/tree/main/skills/review-pipeline
Command: npx skills add https://github.com/Rosetears520/aili-workflows --skill review-pipeline-rosetears520

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents premature acceptance by orchestrating a post-implementation review that reconciles correctness, testing, and security findings into a clear go/no-go decision.

Core Features & Use Cases

  • Targeted Review Orchestration: Collects the review target (diff range, acceptance criteria, prior verification, and unverified items) to avoid irrelevant analysis.
  • Relevant Fan-Out Review: Dispatches only the appropriate read-only reviewer roles (code, tests, security, and optional context mining) based on what matters.
  • Evidence-Based Reconciliation and Fix Loop: Merges duplicate findings, classifies severity, runs a bounded fix/verification loop, and blocks completion unless issues are resolved or explicitly accepted.

Quick Start

Use the review-pipeline skill after finishing a multi-file, behavior-changing implementation and before claiming PASS.

Frequently Asked Questions about review-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent premature acceptance of multi-file code changes before release?

Prevent premature acceptance by orchestrating a post-implementation review pipeline that reconciles code, test, and security findings against evidence before allowing a release gating PASS decision.

What is evidence-based reconciliation in a code review pipeline?

Evidence-based reconciliation merges duplicate findings from dispatched code, test, and security reviewers, classifies severity, and runs a bounded fix loop to resolve blocking items before completion.

How do I run a bounded fix loop for security audit findings?

Run a bounded fix loop by merging duplicate security audit findings, classifying severity, executing narrow verification cycles, and blocking completion unless issues are resolved or explicitly accepted.

When do I need an orchestrated review pipeline for release gating?

You need an orchestrated review pipeline for multi-file changes, security-sensitive work, test-heavy updates, and release-impacting modifications that require coordinated verification and evidence-backed gating.

Does this review pipeline support test validation and security audit dispatching together?

Yes, the review pipeline dispatches relevant read-only reviewer roles for code, tests, and security audits simultaneously based on the review target to avoid irrelevant analysis.

What happens to blocking items in the completion gate of a review pipeline?

The completion gate blocks PASS status unless blocking items identified during findings reconciliation are fully resolved or explicitly accepted through the bounded fix loop verification.